Skip to content

GitLab

Commit 7d0dc183 authored by syiming's avatar syiming
Browse files

remove min_depth and depth_multiplier feature

parent 1538da90
Show whitespace changes
Inline Side-by-side
Showing with 1 addition and 30 deletions
research/object_detection/models/faster_rcnn_resnet_v1_fpn_keras_feature_extractor.py
View file @ 7d0dc183
......@@ -42,8 +42,6 @@ class FasterRCNNResnetV1FpnKerasFeatureExtractor(
resnet_v1_base_model_name,
first_stage_features_stride,
conv_hyperparams,
min_depth,
depth_multiplier,
batch_norm_trainable=False,
weight_decay=0.0,
fpn_min_level=2,
......@@ -61,9 +59,6 @@ class FasterRCNNResnetV1FpnKerasFeatureExtractor(
conv_hyperparameters: a `hyperparams_builder.KerasLayerHyperparams` object
containing convolution hyperparameters for the layers added on top of
the base feature extractor.
min_depth: Minimum number of filters in the convolutional layers.
depth_multiplier: The depth multiplier to modify the number of filters
in the convolutional layers.
batch_norm_trainable: See base class.
weight_decay: See base class.
fpn_min_level: the highest resolution feature map to use in FPN. The valid
......@@ -94,8 +89,6 @@ class FasterRCNNResnetV1FpnKerasFeatureExtractor(
self._resnet_v1_base_model = resnet_v1_base_model
self._resnet_v1_base_model_name = resnet_v1_base_model_name
self._conv_hyperparams = conv_hyperparams
self._min_depth = min_depth
self._depth_multiplier = depth_multiplier
self._fpn_min_level = fpn_min_level
self._fpn_max_level = fpn_max_level
self._additional_layer_depth = additional_layer_depth
......@@ -152,8 +145,6 @@ class FasterRCNNResnetV1FpnKerasFeatureExtractor(
conv_hyperparams=(self._conv_hyperparams
if self._override_base_feature_extractor_hyperparams
else None),
min_depth=self._min_depth,
depth_multiplier=self._depth_multiplier,
classes=None,
weights=None,
include_top=False)
......@@ -166,14 +157,12 @@ class FasterRCNNResnetV1FpnKerasFeatureExtractor(
backbone_outputs = self.classification_backbone(full_resnet_v1_model.inputs)
# construct FPN feature generator
self._depth_fn = lambda d: max(
int(d * self._depth_multiplier), self._min_depth)
self._base_fpn_max_level = min(self._fpn_max_level, 5)
self._num_levels = self._base_fpn_max_level + 1 - self._fpn_min_level
self._fpn_features_generator = (
feature_map_generators.KerasFpnTopDownFeatureMaps(
num_levels=self._num_levels,
depth=self._depth_fn(self._additional_layer_depth),
depth=self._additional_layer_depth,
is_training=self._is_training,
conv_hyperparams=self._conv_hyperparams,
freeze_batchnorm=self._freeze_batchnorm,
......@@ -232,8 +221,6 @@ class FasterRCNNResnet50FpnKerasFeatureExtractor(
is_training,
first_stage_features_stride=16,
conv_hyperparams=None,
min_depth=16,
depth_multiplier=1,
batch_norm_trainable=False,
weight_decay=0.0,
fpn_min_level=2,
......@@ -246,8 +233,6 @@ class FasterRCNNResnet50FpnKerasFeatureExtractor(
is_training: See base class.
first_stage_features_stride: See base class.
conv_hyperparams: See base class.
min_depth: See base class.
depth_multiplier: See base class.
batch_norm_trainable: See base class.
weight_decay: See base class.
fpn_min_level: See base class.
......@@ -259,8 +244,6 @@ class FasterRCNNResnet50FpnKerasFeatureExtractor(
is_training=is_training,
first_stage_features_stride=first_stage_features_stride,
conv_hyperparams=conv_hyperparams,
min_depth=min_depth,
depth_multiplier=depth_multiplier,
resnet_v1_base_model=resnet_v1.resnet_v1_50,
resnet_v1_base_model_name='resnet_v1_50',
batch_norm_trainable=batch_norm_trainable,
......@@ -278,8 +261,6 @@ class FasterRCNNResnet101FpnKerasFeatureExtractor(
is_training,
first_stage_features_stride=16,
conv_hyperparams=None,
min_depth=16,
depth_multiplier=1,
batch_norm_trainable=False,
weight_decay=0.0,
fpn_min_level=2,
......@@ -292,8 +273,6 @@ class FasterRCNNResnet101FpnKerasFeatureExtractor(
is_training: See base class.
first_stage_features_stride: See base class.
conv_hyperparams: See base class.
min_depth: See base class.
depth_multiplier: See base class.
batch_norm_trainable: See base class.
weight_decay: See base class.
fpn_min_level: See base class.
......@@ -305,8 +284,6 @@ class FasterRCNNResnet101FpnKerasFeatureExtractor(
is_training=is_training,
first_stage_features_stride=first_stage_features_stride,
conv_hyperparams=conv_hyperparams,
min_depth=min_depth,
depth_multiplier=depth_multiplier,
resnet_v1_base_model=resnet_v1.resnet_v1_101,
resnet_v1_base_model_name='resnet_v1_101',
batch_norm_trainable=batch_norm_trainable,
......@@ -325,8 +302,6 @@ class FasterRCNNResnet152FpnKerasFeatureExtractor(
is_training,
first_stage_features_stride=16,
conv_hyperparams=None,
min_depth=16,
depth_multiplier=1,
batch_norm_trainable=False,
weight_decay=0.0,
fpn_min_level=2,
......@@ -339,8 +314,6 @@ class FasterRCNNResnet152FpnKerasFeatureExtractor(
is_training: See base class.
first_stage_features_stride: See base class.
conv_hyperparams: See base class.
min_depth: See base class.
depth_multiplier: See base class.
batch_norm_trainable: See base class.
weight_decay: See base class.
fpn_min_level: See base class.
......@@ -352,8 +325,6 @@ class FasterRCNNResnet152FpnKerasFeatureExtractor(
is_training=is_training,
first_stage_features_stride=first_stage_features_stride,
conv_hyperparams=conv_hyperparams,
min_depth=min_depth,
depth_multiplier=depth_multiplier,
resnet_v1_base_model=resnet_v1.resnet_v1_152,
resnet_v1_base_model_name='resnet_v1_152',
batch_norm_trainable=batch_norm_trainable,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment